Supporting Material is available for this work. For more information, follow the link from
the Table of Contents to "Accessing Supporting Material".
Table of Contents
Dynamic Interactions Between Citizen Choice and
Preferences and Public Policy Initiatives
A System Dynamics Model of Recycling Dynamics in a Typical Swiss
Locality
‘The outcome of recycling initiatives depends on both citizens’ policy compliance and on the
development in the recycling markets. The task of localities is to find incentives to motivate
Citizens to participate in recycling programs producing a high quality recycling material.
However, different local policy ietiatne stn Badiewtaad enntn New Vere State showed
some undesired consequences. A System Dynamics model is proposed in order to analyze
the undesired effects and to test further policies. The model design is based ona feedback
theory about human behavior and public policy that stresses both the importance of
contextual and personal factors. Hidden attitudinal stocks in the system create adaptation
delays, leading to unexpected system behavior. A plausible personal structure is suggested
that represents the overall propensity to separate waste. First policy runs show that a
combination of interventions altering personal and contextual factors is superior to single
focused strategies.
Key words: System dynamics, feedback systems, citizen choice and preferences, human
behavior, propensity, public policy analysis, solid waste management, computer: aided
decision making
Introduction
The role of local authorities in public policy initiatives
Local authorities are often in charge of implementing national policies. The success of a
national policy depends on a farsighted development of local ies to motivate citizens
to comply with public policy initiatives, such as recycling programs’. However, the
outcome of local public policy initiatives depends not only on the compliance of the
citizens but also on market forces. Therefore the present paper suggests a System Dynamics
model for municipal solid waste management (the SWM- model) that addresses both long-
term feedback effects of market forces and dynamics of citizen's choice and preferences
that affect the outcome of local and national policies.
Fora theory-based discussion of the role of local authorities see Oates, 1990.
The long-term feedback perspective should help to recognize and analyze important loops
causing desired and undesired outcomes of national and local solid waste policy (see
Richardson and Pugh 1981).
lesired ment Nation plan i ental
piece ance jal to correct undesired development "etic rtenl ae
and management (outcome) in solid waste generation (impact x) >p (output y)
and management (output x)
Development in the free market industry
i Consumption.
' ehaviorof households |
+ Gmpact y) i
Figure 1: Long-term feedback effect of national and local municipal solid waste management policies
Background of the research: The Swiss Priority Program ,Environment”
In 1992 the Swiss National Science Foundation launched a broad program to promote the
research on relevant topics for a sustainable development, the Swiss Priority Program
“Environment” (SPPE*). 500 researchers in 200 projects were underway to analyze and
understand complex systems in order to find solutions to prevailing environmental and
development problems. Especially interesting for the proposed work are the new insights
about environmentally significant behavior of individuals or social groups and policy
building’, such as in-depth knowledge about hurdles and keys to promote environmentally
sound behavior (see Gessner 1996, Gessner and Bruppacher 1999, Kaufmann- Hayoz and
Gutscher 2001).
Households as consumers have an important influence on the success or failure of specific
products and the development of whole industries. Their lifestyles determine what kind of
technologies, materials, and resources will be used. Therefore at the local level households
are important players towards a global sustainable development. However, many studies
about environmentally relevant behavior show that even environmental concemed
consumers face many obstacles when trying to adopt a environmentally sound lifestyle.
Therefore, local interventions that help to overcome those obstacles can be crucial. For
policymakers it is important to have a heuristic that helps to understand the processes and
? http://www.snf.ch/SPP_Umwelt/overview.html
° Particularly the Integrated Project “Strategies and instruments for sustainable development: Bases and
evaluation of applications, with special regard to the municipality level (Nr. 5001-48826) with several
subprojects, (1997 - 2000) and earlier on the Module 4 “Environmental awareness and activity (1992-1996).
factors that influence citizens’ choice and preferences as well as the dynamics that alter the
state of the social system. Subsequently it can help to find the right intervention points.
A first synthesis of findings that emerged from the intensive research about human
behavior and environmental policy instruments has resulted in a simple feedback theory
about human behavior and public policy (see Kaufmann- Hayoz and Gutscher 2001a,
Kaufmann- Hayoz, Battig et al. 2001).
Figure 2 represents the basic assumption about human action as a result of an interaction
between the intemal structure of the actor, (including personal factors) and the extemal
structure (representing contextual factors in the cultural, socio- economic, institutional, and
physical framework). The extemal structures offer options but also constrain human.
behavior. It is a result of a multistage decision process in political/administrative,
technological and economic domains. “All actors have only limited possibilities to alter
their own framework of actions, because they are determined by other actors’ decisions.
However, collective actions or social practices stabilize and reproduce the mutual
framework conditions, or, altematively, they contribute to their change. Over time there is a
‘co-evolution’ of individual and collective pattems of behavior and its framework”
(Kaufmann Hayoz and Gutscher 2001a:24). This feedback view is in line with the control
theory proposed by Powers’ major work, ‘Behavior: The Control of Perception’ (Powers
1973 1990). He emphasizes that individuals not only behave as they do because of the
stimuli they perceive but also that how individuals behave affects what they perceive.
Figure 2: A simple feedback theory about human behavior and public policy
(Kaufmann-Hayoz etal. 2001:82)
(see end of paper)
Based on the feedback theory, Kaufmann- Hayoz et al. (2001) conclude that policy makers
either can alter the intemal or the extemal structure in order to induce behavior change.
Different types of policy instruments such as “command and control instruments”,
“economic instruments”, “service and infrastructure instruments” and “collaborative
agreements” can alter the extemal structure. Policy types such as “communication and
diffusion instruments” and also “collaborative agreements” can modify the individual’ s
intemal structure. A coherent explanation of the different types of policy instruments is
given in (Kaufmann- Hayoz, Battig et al. 2001).
One aim of the present work is to operationalize this broad theory in the SWM-model in
order to gain a better understanding about the dynamics of contextual and personal factors
and possible intervention points. The model should help to gain insights into dynamic
interactions between citizen choice and preferences and public policy initiatives. The focus
is on intended and unintended effects of different local recycling initiatives that result in
policy resistance.
This paper contains three main parts, which are the model conceptualization, the SWM-
model description, and the model behavior. Those are framed by an introductory chapter
and by some thoughts about first insights and conclusions. Since the modeling process is
still ongoing, only the first important model parts and preliminary policy experiments are
described. However, the modeling project will be finished in the summer of 2003 and an
updated version of the model as well as final insights will be presented at the conference.
Issue: Solid waste managementin Switzerland
This study deals with local solid waste management policies in a typical Swiss locality. The
overall purpose of the modeling project is to gain a better understanding of local solid
waste management problems. However, such local problems seem also to be crucial for
national policies, since they address national and global behavior trends (see Duggan 2002;
OECD 2000 in Ludwig, Hellweg et al. 2003). The assessment of the actual situation in
solid waste management made by ‘the Swiss A gency for the Environment, Forest and
Landscape’, can be summarized as follows.
The overall achievement of the solid waste policy is a well- organized management of solid
waste. But the whole production and recycling process still involves an inordinate
consumption of energy and materials. The aim should be to create incentives for waste
avoidance and recycling by levying charges for disposal services, and to promote the
purchase of long-life products. Sustainable use of raw materials and the search for
incentives that promote the necessary changes in behavior are likely to become the key
issues for waste management. Waste mountains are a sign of inappropriate production
methods and behavior (see SAEFL 2002).
A comparison between Switzerland and the USA shows that the solid waste management
system in Switzerland is relatively advanced.
Switzerland’ USA
‘Amount of solid waste 1.2 kg/person/day kg/person/day
2.65 pounds/person/day 4.5 pounds/person/day
Recycled a3 30.1%
Incinerated 57% 147%
Land filed 0% (but 0.64Mio t combusting ash) 55.3%
Table 1: Municipal Solid Waste management in Switzerland and USA (2000)
In Switzerland municipal authorities are responsible for solid waste management. The law®
for solid waste management requires that the services be paid according to the polluter-
pays- principle. Therefore, many municipalities levy charges per collected garbage- bag
(garbage-bag charges). That means that the households are required to pay a charge per bag
for bumable solid waste. Most other collecting-services for recyclable products (paper,
cardboard, glass, ferrous metal - like tins, hazardous waste - like oil, batteries, pet’-plastic,
aluminum, batteries, food scraps) are free of charge; respectively it is intended that the cost
“ http/www.umwelt-schweiz.ch/buwal/eng/medien/umweltbericht/druck/index.html (Swiss A gency for the
Environment, Forest and Landscape)
> www.epa.gov /epaoswer/non-hw/muncpl/facts.htm(US Environmental Protection A gency)
© Environment Protection Law amended 1997 by the Swiss national goverment
Tpolyethylenterephtalat
be covered by basic taxes or by some prepaid taxes. However, the households have to pay
an extra price for the disposal of some recyclable material, (e.g. for metal, electrical and
electronic equipment and appliances).
As aconsequence of the introduction of the garbage- bag-charge- policy, the fraction of
bumable waste decreased and the fraction of recyclable matenal increased. The cost for
collecting the different streams of recyclable material increased as well. Thus the municipal
budget for solid waste management is growing. There is also a need for monitoring the
disposal behavior of citizens, which adds to the cost. Due to higher operative cost and bad
recycling-market conditions the relative profit from delivering the separated material to
recycling instead to incineration plants is decreasing. The households resist paying higher
prices for solid waste management services. This is a critical problem (see Joos, Carabias et
al. 2002) . Disposal services are not perceived as a cost- effective public service and citizens
still expect to get those services for free’.
In order to disburden the municipality from the high cost for solid waste management and
to promote recycling further, the national govemment discusses the initiation of prepaid
disposal charges on more recyclable products. “Advance disposal? charges make it possible
to apply the polluter- pays principle in financing the comprehensive network of take- back!°
points that is required for a high recovery rate, as well as transport and, finally,
environmentally sound processing” (SAEFL 2000:115).
This statement describes the intended effect of the prepaid disposal policy. But there will
probably be some unintended effects, as the prepaid disposal charges give different signals
to the households and to the recycling sector. It may happen that the awareness of the
prepaid price for disposal will decline, and people may put more recyclable material into
the bumable garbage. A further side effect could be that the infrastructure for collecting the
material could deteriorate. Since there are few incentives for retailers, they are not
interested in promoting good collecting services.
® http://www.umwelt-schweiz.ch/buwal/de/medien/umwelt/2002_3/index.html (p35).
® Advance disposal charges are prepaid disposal charges: the disposal price will be included in the product
rice.
° Take-back points are collecting points for recovered recycling material.
1987-
Separated Materials 1991,
1994- 1998-
1997 2001
| 1992 | 1993
Ferrous metal (tins)
Food scraps
Hazardous waste (oil)
Electronic waste
Number of recycling streams [5 [6
Table 2: Recycling streams separately collected by Swiss municipal authorities
A short glance at recycling programs in some localities in New Y ork State shows that they
face similar problems. Duggan (2002) reported recently the actual situatio in the Times
Union. Since there is not yet a strong market for recyclable materials, the cities have to pay
someone to take the separated material for most collecting categories. Furthermore the
operating expense for picking it up separately generates higher costs than if it would be
dealt as normal “trash”. Hence “recycling” can become economically questionable for the
localities. The following two voices of local solid waste management experts point out two
major problems they have to deal with. B. Chamberlain, Troy’s solid waste management
coordinator observes, “If the secondary markets don’t improve, the prices to recycle certain
material will go up, and once it passes what it costs to landfill it, it won’ t be economically
beneficial” (cited in Duggan 2002). A further important aspect of recycling programs deals
with the quality of the separated recyclable material. In the case of New Y ork City they
created a useless mixed material since they collected and compacted glass and plastic
together, for which there is no demand. J. Enck, a policy advisor stated: “State law does not
prescribe how you are supposed to do the collection. Y ou can sabotage a recycling program
if you wanted to” (cited in Duggan 2002).
These developments in the real world set the stage for the proposed research. The SWM-
model should help to address and analyze these observed problems.
Model conceptualization
Problem statement
The problem addressed by the System Dynamics model is represented in the following
questions:
What local policies increase recycling, reduce the overall generation of solid waste, and
help to establish / ensure a solid waste management system that fosters competitive
recycling markets?
¢ How do you motivate the households to participate in solid waste reduction and
separation?
¢ How do you recover recyclable material to produce competitive secondary raw
material?
¢ How do youfinance the recovering and disposal activities of local agents?
Reference modes and problem dynamics
In the following paragraph, some variables of interest and their historical dynamics from a
typical Swiss locality over the last 14 years (1987 - 2001) will be presented. Subsequently,
the question “What caused the given development?” (see Randers 1996) will be addressed.
Chart 1 shows the development of the municipal budget for solid waste management. There
is an increase in cost over time and in some periods there was a deficit. However, there was
also an increase in the amount of solid waste during this time. Therefore, this chart gives us
no information about the development of costs per kg.
Bugdet: solid waste management
1400000
1200000
1000000 eee
00000 —— expenditure sFr
ws |—"— revenue sFr
600000
ZS eee —— deficit Fr
400000 #
200000
4 Van aa
9 SN DM oY 9 oh 2
0° 0 2 mr 9
BPP” SPP? PP
Chart 1: Municipal budget development for solid waste management
In order to slice the problem (see Saeed 1992) and to decompose the growth trend of solid
waste generation, the budget of solid waste per capita and per kg is computed (see Chart 2).
Budget: solid waste management
(per capita per kg)
0,3
0,25 See
0,2
—#— expenditure per capita per kg per year sFr.
0,15 +>
—+— revenue per capita per kg per year sFr.
0,1 —+— deficit per capita per kg per year sFr [
0,05
————— ae
SP oP PD SY Sh 5? OF 67 OH 5 3% oP
0"
FPF PPP PP SP SPP IPS oS
Chart 2: Budget development per capita per kg per year
According to Chart 2, there is an upward trend in the unit cost that peaks in 1994 followed
by aslight drop, and then it seems to reach a plateau. However the revenue continues to
fall. There are two periods with a higher deficit (1997 - 1992) and (1996-2001). As the
deficit has grown, the local authorities increased the tax for solid waste management and
the volume related trash bag charges (see Table 3).
Time period 1973-1990 | 1991 - 1999 2000 - 22?
Taxes, per year, according to the |24-60sFr. |50-110sFr. | 83- 184sFr.
size of the apartment / house
Volume related trash-bag None 0.9sFr 1.80 sFr.
charges (35 liter)
Table 3: Changes in taxes and trash-bag charges (Einwohnergemeinde X 1973 - 2000), (Einwohnergemeinde
Xa1973 - 2000)
Chart 3 illustrates the changes in the number of recycling streams. Between 1991-93 four
additional recycling streams were offered to the citizens. From 1993 to 2001 only one
additional recycling stream was introduced.
Number of recycling streams
10,0
9,0
a re ae
7,0
6,0 4
5,0
4,0
3,0
2,0
1,0
0,0
s & cal & se cu & s £ x ca x & aS
= number of recycling
streams
Chart 3: Development of number of recycling streams
Chart 4 portrays the change in the fraction of separated material and the material disposed
for buming. The fraction separated for recycling increased from about 30% to 50%.
Historical development fraction separated
0,9
0,8
ee
0,7
‘6 [ \
05 |
04
03 NES traction separated for
: 5 ae en ee recycling of the msw
02 L
B= Fraction put for burning of
O1 the msw r
0,0
Sy ob > ot oO © Sh oO
OY "8? oP o_o oO P Ss
~ PPP PP A PE A A of
Chart 4: Historical development of fraction separated for recycling (GSA , Bauverwaltung Gemeinde X 1997)
Between 1990 and 1992 there was a transition phase with a short term dynamic in the
fraction of separated waste and the fraction disposed for buming. These dynamics can be
ascribed to the implementation of the trash bag charges in 1991. As the citizens leamed that
the disposal cost would increase, they started to clear out useless material. In 1991 the price
incentives had a strong effect on the disposal behavior of the citizens. They probably over
invested in separation activities since they tried to avoid disposal cost (over reaction).
However, the SWM- model will not address these short term effects. Nonetheless, it is
interesting to see that the 100% increase in the trash bag charges had nearly no behavioral
effect in year 2000. A reason for this phenomenon could be that the monetary incentive
givenin 1991 was high enough to activate the potential capacity of citizens to separate
given a constant amount of recyclable material. Stems concept of limiting conditions would
explain this effect on the individual level with diminishing retums of interventions. If the
financial incentives demonstrate a clear personal benefit a further increase may be far less
effective than other interventions (providing more opportunities, giving better information
or other incentives) see (Stern 1999).
To explain the long-term dynamics of the reference modes the following dynamic
hypothesis is postulated.
Since the performance of citizens’ separation behavior was low, the localities gave price
incentives in form of a garbage bag charge. The intended effect was to promote the
separation behavior. As a consequence the fraction of separated waste increased and the
relative amount of solid waste for buming decreased. The unintended effect was that not
only the relative amount of waste disposed for buming decreased. But also the revenue
generated from the trash bag charges declined. Therefore, the budget deficit started to
increase. A further increase in the price for bumable material had nearly no additional
effect on the separation behavior, since the number of recycling streams was held nearly
constant. The citizens had no real legal option to avoid higher costs for disposing the
bumable material. As an unintended consequence, the quality of the separated material
decreased. Citizens started to put bumable material in the recycling streams. However, this
effect was only observed and could not be exactly quantified.
The following causal loop diagram shows the postulated main feedback loops that are
responsible for the dynamics of the variables of interest. The balancing feedback loop
“limiting propensity from time cost” refers to the citizens ‘ behavior (Figure 3) and the
other “deficits limits investments” refers to the authorities (Figure 4).
The balancing feedback loop
“limiting propensity from time cost”
postulates that a high propensity to
separate would foster (with a delay)
the development of further recycling :
streams. This link represents the citizens! cay
theory that a high discipline in tne edst-tise bia
separation behavior of citizens po og
would increase the purity of the (-}
separated recycling material. As a me
consequence, the recycled material
would become competitive,
fostering the development of ee ee
recycling capacity, and new separating, [ citizens:
recycling streams. As the number of Spo senarate
recycling streams increases, the time Figure 3: time cost limiting propensity to separate
cost to separate increases. This (citizens’ choice)
results in both a lower perceived
profit and propensity.
recycling local capacity
time cost for streams for recovering
+
separating ¥,
+ cost for
collecting
relative profit
citizens" from'sw
citizens’ perceived jm propensity management
profit from to separate (authorities)
separating
Figure 4: Deficits limits investment
(choice of the authorities)
The balancing feedback loop “deficit limits investment” describes the economic concems
of the localities. As long as the price for disposing the separated material is lower than for
bumable material, there is a relative profit in the local recycling program. As the number of
recycling streams increases, the operating cost elevates for collecting the various separated
materials. Therefore, the relative profit from the recycling program decreases and the
willingness to invest in local capacity decreases as well.
These two balancing loops indicate that there will be an upper limit in the number of
recycling streams, due to limited local capacities.
Figure 5 captures the pricing structure in Swiss localities that creates a reinforcing feedback
loop “propensity to separate increases deficit’. This loop describes the unintended effect of
a growing deficit between the revenue and expenditure for swm- services in the period from
1996 - 2000, (see also Chart 2). The price incentives given by the trash bag charges
increased the propensity to separate. As a consequence, the amount of material disposed for
buming decreased relative to the amount of separated material, resulting in lower revenue
from bumable material. Therefore, not only the relative but also the overall profit decreases
(respectively the deficit increases). Consequently, the authorities raised the price for
bumable waste in year 2000. This reinforcing loop indicates that this pricing structure will
not ensure a sound solid waste management system.
+
*
a number (gj invest ment in
citizens
recycling local capacity
time cost for streams for recovering
separating +
= deficit limits investment
limiting propensity from time cost
*™ cost for
collecting
- relative profit
citizens” fromew
citizens’ perceived Bt propensity
t solid t
Lp fraction to amoun managemen
Brentitien to separate [be burnt Pmnaste to be Revenue tion (authorities)
g 2 material
th) a
propensity to separate increases deficit generation
A
citizens’ perceived
cost for burnable sw price for
urnable sw
t
Figure 5: propensity to separate increases deficit
The next two reinforcing loops “policy resistance” (Figure 6) explain, how a further
unintended effect sabotages the local recycling program. Due to price incentives the
citizens perceive a high profit from separating (see the lower feedback loop) and their
propensity to separate increases. Since only a limited fraction of solid waste is recyclable,
the citizens are tempted to put bumable material into the recycling streams in order to avoid
disposal costs. Therefore, the impurity in the separated material increases. As a
consequence, the recycling industry is not going to accept these materials or will charge
higher prices. This increases the operating cost of the localities and decreases the relative
profit and also the willingness to invest in local capacity for separating. Therefore the
number of recycling streams could decrease. Given a high propensity to separate, citizens
continue to put bumable waste into the recycling streams.
+ +
number |egj——_ investment in
citizens)
recycling local capacity
time cost for streams for recovering
separating +
= de et we
limiting propensity from time cost
* > cost for
—_
amount, burnable collecting
_ amgunt yennable TL
recycling steam
wee impurity) 4)
= - relative profit
auzes policy resistance een
citizens’ perceived pl
propensity 1ackIaNeS. amount solid management
profit from to separate [—M ve burnt ——Pewaste to be ge revenue from PR uthorities)
separating burned material
? ~
t+)
propensity to separate increases deficit
citizens' perceived
cost for burnable sw price for
rh
Figure 6: policy resistance
The outline of the problem and the dynamic hypothesis give evidence that the number of
recycling streams is an important stock at the local level. The following causal loop
diagram (Figure 7) gives a reason, why the number of recycling streams is also a critical
factor for development of recycling markets.
A higher number of recycling streams decrease the cost for recycling, since the recycling
industry gets a better quality of collected material. Hence it has to invest less in sorting
processes. Lower production cost of secondary raw material increases the profit and
reduces the relative price of recycled raw material. Therefore the demand for recycled raw
material will increase. Furthermore the supply and the variety of recyclable material in
products will increase. As a result of a successful recycling market, not only the willingness
to invest in higher capacity increases but also the readiness of new recycling technologies
to enter the recycling- market increases. Therefore the number of recycling streams grows.
If the citizens will separate the recyclable material according to the different recycling
streams, the recycling industry will face lower recycling cost. In this scenario the
reinforcing feedback loop will foster a growth in the recycling market. Otherwise, higher
processing cost from impure recycling material will shut down the recycling market. These
scenarios will be analyzed in the model.
Results of some pilot- experiments and studies about expanded recycling initiatives for
plastic in different Swiss localities give empirical evidence of the stated dynamic
hypothesis (BUWAL 2001).
+ [number
recycling
streams
supply of different
recyclable material
in products cost to sort out impurity
+ (recycling industry)
4)
rs
trap/chance recycling market
demand for recycled ¢
material (production profit (recycling
industry) - industry)
= +
relative price for
recycled material
Figure 7: Chance for the recycling market
Effects of prepaid disposal charges
With the swm-model not only the given development will be addressed, but also the effect
of further policy- strategies, such as prepaid taxes. The model will be designed to give
insights to the question: what are the likely effects of other strategies such as prepaid
disposal charges on a growing number of products?
Prepaid disposal charges have an important feature. For the consumer this is a hidden price.
Therefore, the collecting service system will have a feedback structure of non- price
mediated resource allocation (see Sterman 2000).
A higher service quality in the collecting centers stimulates the propensity to separate.
Citizens will bring back a higher amount of different recyclable material.
A higher amount of collected aoe
recyclable material erodes the Propensity amen ae deer ee
: ; ; to separate recyclable material
service quality as the crowding #
+
increases. This dynamic represents 4) |
number of
the balancing feedback loop “limit
of recyclable growth”. This means aoe rte ere esac)
that the service quality will limit the quality service request
amount of collected recyclable , ae 2 \ .
material. The second balancing »)
feedback loop, “limit of resources”, =
shows that a higher service request Meieyech
increases the need for adequate scclice veer neve aM ee at
services. The increased adequacy of we
service will demand a better .
infrastructure, which would elevate Figure 8: Feedback structure of the non-price mediated resource
the cost. As a consequence, the Hlloralion sys ees from
availability of service resources will
be diminished, resulting in lower
service quality.
The two balancing loops indicate that a prepaid disposal charge can foster the separation
behavior of citizens only to a certain limit. Once the propensity to separate tends to
decrease, the fraction separated for recycling will stay constant on a certain equilibrium
level, even when the number of recycling streams will increase.
Chart 5 represents the hypothesized reference mode that takes into account the underlying
balancing feedback structure explained in Figure 8.
Due to the balancing feedback loops the fraction separated for recycling will reach an
equilibrium position while the maximal acceptable number of recycling streams for citizens
will be reached. Due to information delays in the market system the number of recycling
streams will increase further resulting in an overshoot in the number of recycling streams.
14
Fraction recyclable of the
msw
124 fraction burnable of the
msw
10 U7 tumber of recyting i
o stream (scale #/10), Pe
pea:
08
7
'
06 if >
sooce!
" nn
0,2 5
Historical data i Hypothesized data
00
SN oY 9? 0 of oy ? @ oh
D> 0” oD? oD GO" HD? PH’?
PP PP PP PAH’ AH
2 wy A> 6 A 09
PMY Ww Ww
PP PP
Chart 5: Hypothesized development of fraction separated for recycling and number of recycling streams
For the time frame of this study a steady increase in the overall amount of waste (including
both bumable and recyclable waste) is hypothesized, which could be even exponential. This
assumption reflects the observation that solid waste generation is highly correlated with
economic growth!!, The scenario of economic growth shows the extemally driven behavior
pattem. This component of the behavior pattem is modeled in a smaller subsystem that can
be switched off. Subsequently, the discussed developments in the SWM- model can be
analyzed either with or without economic growth - scenarios. This helps to partition the
messy problem in the solid waste management into macro- economically driven
developments and into policy- incentive driven developments (see Saeed 1992 in
Richardson 1996). One additional challenge of this work would be, to analyze if the macro-
economically driven development could he influenced by local policy interventions. Under
which condition could a growing green consumerism result in solid waste avoiding
behavior (see also Joos, Carabias et al. 2002)!
1 http://www _umwelt-schweiz.ch/buwal/eng/medien/umweltbericht/druck/index.html (Swiss A gency for the
Environment, Forest and Landscape)
12 http://www.IP-Waste.unibe.ch/public/A bschlussband/inhaltsverzeichnis.html
Purpose of modeling
The model is designed to create a computer based leaming environment or a micro world
for local policymakers to play with their knowledge of the solid waste system and to debate
policy and strategy change (see Morecroft 1988). It can be used ase a communication tool
to enhance a debate between the different agents about organizational structures in the area
of solid waste management (see Schwaninger 1997). Finally it adds to the scientific
discussion about long term dynamics between citizen choice and preferences and public
policy initiatives.
To be more concrete, the following objectives should be met:
Firstly, the model help to discover the underlying causes of changes in the fraction
separated and the quality of the separated matenial.
Secondly, the model is designed to uncover and clarify possible side effects of changes in
the price structure and of prepaid disposal charges.
Lastly, the model helps local authorities dealing with mandates from the federal
govemment and implementing sound solid waste management policies.
The overall model structure
In order to analyze long- tem effects of different local policy intervention a time horizon
from 1987 to 2020 was chosen. For the time period 1987 to 2001 there is data available (see
reference modes) revealing historical pattems of behavior. The time span of two decades
from 2002 to 2020 allows experimenting with further policy options and strategies and
analyzing their behavioral impact. The sectors in the model are seen from a specific
distance in order to see the intemal structure, social pressures, market forces, and important
decision points. A balance between a microscopic view that is too psychological and a
telescopic view that captures an economic perspective that is too aggregated is aimed for
(see Forrester 1961, Richardson 1991). Therefore in the model the different recyclable
materials will be aggregated to one flow. However, the model is designed to focus on the
number of different recycling streams and the effects of a change in the number.
The solid waste model includes the following sectors (see Figure 9):
The main sector is the local separation sector that is disaggregated in the following sub
sectors: the household waste separation sector, the household decision sector and the local
solid waste management sector. These sectors include endogenously operating dynamics
deemed important to address the solid waste management problems and to conduct policy
analysis.
The household waste separation sector includes:
¢ The different flows and qualities of the bumable and recyclable waste that result
from separation activities of different groups of citizens.
¢ The initial amounts of different waste qualities, and recyclable and bumable
material will be given exogenously but will be modified by behavioral effects.
¢ The habits of different groups of people to dispose their waste and factors that lead
to changes in habits (i.e. changes in relative prices and the number of recycling
streams).
The household decision sector will describe:
¢ What factors influence the decision of people to become willing / unwilling to
separate the recyclable material?
¢ What influences the willingness to spend time or money on waste separation
activities?
The local policy sector / solid waste management sector includes:
* The development of municipal budget for solid waste management under different
policy options.
* Capacity building processes and the effect of a backlog of separated waste.
Furthermore basic structures of the recycling sector, the supply sector and the incineration
sector are designed in a higher aggregation representing the development of recycling
markets. In these sectors capacities, prices and changes in number of recycling streams will
be computed. This information will be transmitted into the local separation sector.
Some aggregated information about the impact on the environment of incineration and
recycling activities and of the exploitation of raw material from the supply sector will feed
in the household decision sector. The income per capita and the population are given
exogenously. Some time delays due to unavailable and delayed information will occur at
different decision points such as in capacity adjustment processes influencing the system
behavior (see Chung 1992).
Figure 9: Overall model structure
(see end of paper)
Possible local policy options
For forecasting, different policy strategies will be designed that are characterized by specific
“bundles” of parameter values.
“Business as usual policy’
The “business as usual policy” represents the actual policy and is simulated in the base run. This
scenario forecasts the development of the amount of waste and the solid waste management budget
in the municipality without new interventions. This scenario constitutes the base run. The prices and
the number of recycling streams will be held constant after 2000.
“Environmentally ignorant policy”
In this strategy, the main goal is to minimize the disposal cost. No special incentives for households
are given. The specific parameters are: hidden prices for recycling services, lower service quality
resulting in higher time costs for separating, constant or fewer number of recycling streams, lower
garbage bag prices, lower budget for solid waste management.
“Separating policy”
The “separating policy” aims to offer convenient (time saving) collecting services for households
resulting in a higher budget for solid waste management. A high number of recycling streams is
offered in order to collect a high quality of recyclable material. There is probably a trade off
between minimizing the time for separation behavior and enlarging the number of collecting
streams.
“Waste avoidance policy”
In this strategy, the local authorities aim to show the real cost for collecting services and they would
spend more money into educational programs. This strategy implies the following parameter
changes: higher budget for solid waste management, more investment in local service capacity, a
transparent price structure (no taxes), prices for collecting separated material. Probably this strategy
would show a “first worse before better” behavior.
The Solid Waste Management - model
This section firstly describes existing model parts in depth and secondly gives an overview of model
parts that are still under construction. Since the concept of propensity - the propensity of citizens to
separate - seems to be crucial for the success of recycling programs, it will be modeled explicitly.
Therefore a special weight is put on the formulation of the decision process guiding citizens’
behavior to separate.
In the feedback theory about human behavior and public policy (Kaufmann- Hayoz, Battig et al.
2001), contextual and personal factors in a decision making process are emphasized. Therefore in
the SWM- model, interactions between contextual and personal factors will be addressed. Hidden
attitudinal stocks in the system can create adaptation delays leading to unexpected system behavior
and unintended consequences. Elements of the attitudinal structure will be represented in the
household decision and the household separation sector.
Designing propensity to separate: The household decision sector
Citizens’ disposal behavior is seen as a routine behavior and not as a planned behavior. In
Fomrester's term this would be called an informal policy. “... But most guiding policies are informal,
although fully as influential. Informal policy results from habit, conformity, social pressure,
ingrained concepts of goals, awareness of power centers within the organization, and personal
interest” (Forrester 1994:58).
This assumption suggests that people decide once whether to separate or not. Once they have made
this decision, they set a new routine, resulting in new separating habits (see also Dahlstrand and Biel
1997). This implies that there are two main groups of citizens: a group of people “willing to
separate” and a group of people “not willing to separate’. However, in each population we can
distinguish sub groups that are transients (see Figure 10):
* Inthe group “people willing to separate” there are some inexperienced people - they will
show a lower separation performance than the experienced ones. But as they leam to
separate they will move into the stock “experienced people”. The “time to leam” determines
how long this takes.
« Inthe group “people not willing to separate” there are experienced people that got
disappointed from separation consequences. The “experienced people not willing to
separate” will move into the stock “inexperienced people not willing to separate” as they will
forget, they are changing their separation behavior and set up a simpler routine behavior. The
“time to forget’ calculates when these people will move on.
Figure 10: Changes in citizen’ s willingness to separate
(see end of paper)
The flow between the two groups of people “willing to separate” and “not willing to separate” is an
important decision point in the system. Therefore, its decision rule determining the rate has to be
precisely determined (Forrester 1961, 1994, Sterman 2000). The goal would be to formulate the
decision rules with sufficient accuracy in order to gain insight into how people respond to different
circumstances, pressures and policy interventions. Following Forrester (1994) the aim is not to
mimic a process of planned behavior. He suggests a process of seeing goveming policies rather than
individual decisions.
The decision to separate is influenced by the social norm to separate, the time cost of separating, and
teal cost for separating. The decision to become unwilling is influenced by altemative cost such as
time cost and real cost for buming and the social norm for buming. In a further advanced version of
the model, factors such as “perceived policy effectiveness” and “knowledge” will be included in the
decision function. The information about the decision cues (e.g. time cost, real cost, later on the
perceived policy effectivness) comes from other model sectors. In the following part some
psychological assumptions are described.
Some psychological assumptions
The decision rule applied is based on some psychologically grounded assumptions (Latané 1981,
Cialdini, Reno et al. 1990, Hopper and Carl-Niesen 1991, Reno, Cialdini et al. 1993, Mosler 2000,
Mosler, Gutscher et al. 1996, Black, Stem et al. 1985). In the following paragraph they will be made
explicit and explained.
Social norm. baserun
The perceived social norm for
separating is a function of the .
fraction of people “willing to fi sn ae
separate”. An increasing fraction tom atin norm
of people “willing to separate” in Separaling 0.2
the municipality, will generate a
stronger norm to separate,
resulting in a higher number of
people “willing to separate”. In
the decision function this idea is
represented ina non-linear Normal fractional rate,
iia dey a: Given the , becoming willing"
~ (0.1 per year)
environmental problems” it is 0.1 ai
reasonable to assume that a 0 0.50 1
small “normal” fraction of Perceived fraction social norm separating
people will become willing to
separate even when they Chart 6: Fractional rate from social norm separating
perceive no or only a minimal
social norm to do so.
The nomal fractional rate “people becoming willing’ is assumed to be 10% per year, resulting in a
doubling time of 6.93 years!” reflecting the maximal diffusion delay (ceteris paribus). The fractional
rate will increase when nearly 50% of the population generates a social norm to separate. When
nearly all the population is willing, a maximal fractional rate (0.2) will be reached. This value
computes the minimal diffusion delay in the population (doupling time 3,5 years).
The s- shape of the relationship reflects the assumption that first people that are easy to convince will
become willing and later on those that are harder to convince.
Maximal fractional rate
»becoming willing"
(0.2 per year)
a Doubling time = In(2)/fractional rate “becoming willing” (see Sterman, 2000:269)
Acceptable time to separate
The wifingness si spend ane for baserun
separating is a ion of the 7 ‘
aveived social noo to-saparat z acceptable separating time
Itis assumed that people have a 1
maximal acceptable time, they are
willing to invest in separating 0.75
activities. However, this time
would be lower, if the social norm 0.5
to separate is low. Chart 7 shows 0.25
this relationship and discounts the .
maximal acceptable time (y- axis) 0
when the social nom to separate 0 0.50 1
goes down (x-axis). x
Acceptable separating cost -X-
The graph “acceptable separating Chart 7: Acceptable time to separate
time” is based on the same (X-axis: perceived social norm to separate, Y-axis: discount in percent of the
assumption as the concept of maximal acceptable time)
acceptable time. The maximal
recycling cost that people are
willing to pay, will be discounted,
as the social norm to separate will
decrease.
Effect of time cost separating
The graphical converter “z effect of
time cost separating” computes the z effect of time cost separating
effect of the time cost on the
diffusion process (see Chart 8). The 2
effect of time cost is normalized;
when “time spent for separating” = 1,5
“acceptable time for separating”; the «ae Be
graphical function passes the 1 4 afiffush
reference point (1,1). If the time
cost is very low the diffusion
process will be accelerated to a
maximal value of 1.5. If the
required time spent for separating 0 i
(TSS) is twice as high as the 0 1 2
acceptable time for separating TSs
(ATS) the diffusion process will be
stopped. ATS
Chart 8: effect of time cost separating
No effect on diffusion
from time cost (1,1)
To high time cost
will shut down the
diffusion process
Effect of separating cost
The graphical converter “z effect of separation cost’ would calculate the effect of some prices for
separating services in a similar way as the converter for “z effect of time cost separating” described
above.
Decision rules
The “fractional rate becoming unwilling” is formulated in a similar way as the “fractional rate
becoming willing’, but the altemative buming and time costs and a social norm to bum will
determine the rate. A multiplicative formulation of the three decision rules “fractional rate from
social norm”, “effect of time cost separating”, and “effect of separation cost” will be used since any
extreme value in each of them can dominate the other effects as well as one effect can also reinforce
another.
Fractional rate becoming willing = “fractional rate from social norm” * “effect of time cost separating” * “effect of
separating cost”
However it is assumed that the two stocks “experienced people willing to separate” and
“inexperienced people not willing to separate” will never get to zero. There will always be a fraction
that will not change its behavior. This design would represent people with strong beliefs, people that
just do not see any profit, or that are over occupied by separating.
The household waste separation sector
In the household waste separation sector, four different qualities of waste will be computed. The
waste generated consists of recyclable material (A- waste) and non- recyclable material (B- waste).
Therefore, the people have four different action choices to dispose the waste (see Figure 11).
A: The recyclable material can be appropriately separated (A 1) or can be disposed for buming (A 2).
B: The non recyclable material can be disposed for buming (B1) or it can be inappropriately
separated (B2) (generating impure and more expensive recycling material).
Figure 11 explains, how the different qualities of waste are computed.
Figure 11: Action choices for disposing the waste (wep: willing experienced people)
(see end of paper)
The per capita waste generation for all four groups is
assumed to be the same over the years and will be
. Waste per capita 339,0 kg/person/year
held constant: 339 kg/person/year (based on real data Waste put for incineration 247,6 kg/person/year
1987, Table 4). - 65 Waste separated 91,3 kg/person/year
The real data of the different waste qualities “waste
put for incineration” and “waste separated” reflect an Table 4: real data 1987
average system performance and a mixture of A land
B2, respectively A2 and B1 waste qualities.
However in the model it is assumed that the four different groups of people have different disposal
habits, generating different amounts of the four waste qualities.
Chart 9 illustrates the
assumed waste composition
of the four groups of people.
The compositions are
calibrated, based on data of
generated waste per capita in
1987.
Given the disposal habits of
the four groups, their
contribution to four qualities
can be shown. The
“inexperienced people not
willing to separate” start to
produce 100% of the
inappropmiately separated
waste (see Chart 10).
Intial values 1987
‘AL appropriately separated
‘AZ recycable disposed for buming BL norecylable disposed for
‘uring
2inappropaely separated
Chart 9: Waste composition of the four groups of people (Initial values assumed
100%
for 1987).
Values baserun 1992
na
4a
13
no
The different amounts of each group and
quality are added together and the fraction
separated can be computed (Table 5). The
model is calibrated to the real data in 1987.
willing to separate” are influenced by changes in the number of recycling streams.
‘ALappropriately separated AZ recye
Model output 1987
Chart 10: Contribution of the four groups to the different qualities of waste (model
data 1992 baserun)
Initial amount
of waste
Total amount solid waste
3 628 M kg/year
Total amount disposed for burning
2 631 M kg/year
Total amount separated
997 W kgiyear
Table 5: Model output 1987 calibrated to the real data 1987
As the people move from one group to the other the total amount of separated material will change.
The disposal habits of the group “inexperienced people not willing to separate” are influenced by the
relative price buming cost to separating cost. The separation habits of the “experienced people
Local separation sector
In the following section an overview of the three main capacity building sectors will be given,
including the local separation sector, the recycling sector and the production / supply sector. This
model partis still under construction (see Figure 12).
In an abstract sense, the average propensity to separate can be seen as the capacity of the citizens to
separate the recyclable material. This capacity and the local capacity to collect the separated waste
will determine the flow “separating recyclable material by households”. Three feedback loops will
determine the rate “separating recyclable material by households”. Ass the backlog “separated
recyclable material in localities waiting to be recycled” increases both rates “capacity building for
separating” and “separating recyclable by households” shut down. However an increasing demand
of separated recyclable material promote the capacity building process. In all of the three different
feedback processes there are both information and capacity adjustment delays, leading to an unstable
system behavior and inefficiencies.
Recycling sector
The same underlying system-structure affects the rates of flow in the recycling sector. As the
backlog “recycled raw material waiting to be tumed into goods” increases, less material will be
recycled and less recycling capacity will be built. However, an increase in demand for recycled
material will increase the “capacity building for recycling”. As a consequence of an increase in
capacity building, the number of recycling streams will increase, too.
The Production and Supply Sector
In this sector again, the same system structure will be modeled. The backlog “recyclable material in
goods waiting to be separated” wil be computed by the average amount “recyclable disposed for
buming’ (A2- waste from the separation behavior sector). A higher demand of recyclable material in
products by households increases the capacity for processing recycled material. The “actual amount
recyclable material” computed in the household sector will measure the “demand of recyclable
material in products by households”.
This overview of the model structure clarifies the hypothesized reinforcing feedback loop
“trap/chance recycling market” presented in Figure 7. Furthermore, it explains the link of the “local
separation sector” to “the recycling sector’ and “supply sector” determining the development of
recycling markets. Price signals and the perceptions of backlogs will adjust the capacity building
process in all three sectors. Different capacity development scenarios will be simulated. It is
expected that delays lead to undesired effects such as over- investments in capacity building in the
different sectors. Furthermore this structure should also help to understand the trade off between the
maximal capacity to separate of citizens and the capacity development in the recycling sector.
Figure 12: Conceptual overview: Effects of Demand and Backlogs on capacity developments
(see end of paper)
Model behavior
First simulation runs show the dynamic of the model- structure representing the propensity to
separate. The base run describes the model behavior with the actual policies in place: an increase in
number of recycling streams and an increase in the price for a garbage bag in 1991 and 2000 - (In
Graph 1 and 2 the historical data are adjusted to a three median smooth). The simulated fraction
separated and bumed closely tracks the smoothed real data (See Chart 11 A). There is a clear trend
of gowth in the fraction separated. Based on the historical growth trend the model data forecast a
further increase in the fraction separated till it seeks equilibrium at 53%.
REFERENCE FRACTION SEPARATED
1987 1990 1993 1996 1999 2002 2005 2008 2011 2014 2017 2020
Time (year)
three median smooth fraction burned : baserun
Dmal
three median smooth fraction separated : baserun Dmal
fraction for burning : baserun Dmal
fraction separated : baserun Dmal
Chart 11A: Fraction bumed and separated (Base run)
(Line 1 and 2 are three median smoothed real data)
The dynamics are created by the flow of people respectively by changes in the number of the four
different groups of people willing / not willing to separate. Chart 11 B shows a clear increase in the
number of “experienced people willing to separate” beginning in 1991, and a decrease in the number
of “inexperienced people not willing to separate”.
MOVING PEOPLE
10,000
7,500
5,000
2,500
0 Pe
1987 1993 1999 2005 2011 2017
Time (year)
experienced people willing to separate : base ——————— I People
inexperienced people not willing to separate: baserUn —j—————j—e COPE
inexperienced people willing to separate : baserun 3 33 people
experienced people not willing to separate: baserun 4 4 4} 4 people
Chart 11 B: Number of the four groups willing / not willing to separate (Base run)
The outcome of the current policy is described in tems of “total amount appropriately separated”
and “total amount inappropriately separated”. Chart 11 C illustrates an increasing trend in separated
material. However, the price incentives lead to a sudden increase in the amount of inappropriately
separated waste in 1991 and in 2000. But the decreasing trend in the number of “inexperienced
people not willing to separate” reduces this amount over time to a equilibrium level. The gap
between recyclable material and the appropriately separated material decreases, resulting in a
smaller constant gap.
WASTE SEPARATED
4M kg/year
60,000 kg/year
2M kg/year
30,000 kg/year
0 kg/year
0 kg/year
1987 1993 1999 2005 2011 2017
Time (year)
total amount appropriately separated : baserun —+——3—3-_a-_3_kg/year
total amount recyclable material : baserun. ——~——2—>a 2» > kg/year
total amount inappropriately separated : baserun ——>——3——S—S—>-_ kg/year
Chart 11 C: Total amount separated waste (Base run)
First policy tests
To gain further confidence in the model and to test its relevance, first policy tests were conducted.
Since the base run explains the historical behavior pattem with the policy in place, the model can be
used as a laboratory to address the question: What would have happened if other policies had been
chosen? Three altemative policy experiments - a steady state policy, and two altemative separating
policies - give some insightful answers:
“Steady state policy (do nothing)”
In this strategy, there are no price incentives (no garbage hag prices) and no changes in the number
of recycling streams. Therefore, the amount of recyclable material in the waste does not change over
time.
Charts12 A-C: portray the dynamics of the steady
state policy: the fraction separated stays on a
constant level. Over time only slightly more
people become willing to separate. The total
amount of separated waste stays nearly the same.
Further more there is a remarkable gap between
“total amount appropriately separated” and total
amount recyclable material” indicating alow
policy compliance (Chart 12C).
However this policy result is very sensitive to
parameters changes. The model parameters
operate near a tipping point - that means that
different policy outcome would he possible
depending on which loop dominates the diffusion
process (people getting motivated or people
getting disappointed).
MOVING PEOPLE
1987 Toa THT TOO TOIT TOIT
Time (year)
REFERENCE FRACTION SEPARATED
06 y
o4
1987 1990 1993 1996 1999 2002 2008 2008 2011 2018 2017 2020
Time (year)
Dmal
Daal
Dmal
Daal
Chart 12 A: Fraction bumed and separated (Steady state
policy)
thee
WASTE SEPARATED
2M_kglyear [—J
0.2 kglyear
14M kglyear
0.1 kglyear
800,000 kgiyear
© kegiyear
1987 tg98 99920020120
Time (year)
total amount appropriately separated : no chg in price and streams a kg/year
total amount recyclable material: no chg in price and streams —=—2—>— kglyear
‘otal amount inappropriately separated : no chg in price and streams gemmee Kg/y car
Chart 12 B-C: Impact and outcome of steady state policy“
Giving price incentives”
In this policy only price incentives to separate were given. Since 1991 the people have to pay a price
per garbage bag. In 2000 the price increased by nearly 100%. The amount of recyclable material in
the waste remains constant.
According to Charts 13 A-C, the policy increases
the fraction separated slightly till it seeks
equilibrium around 38%. Over time nearly all the
people become willing to separate. The total
amount separated finds equilibrium at a higher
level. Due to the price incentives, the amount
inappropriately separated peaks around 1991 and
2000 and decreases gradually to a low stable
level.
Furthermore, the gap between the recyclable
material and the appropriately separated material
REFERENCE FRACTION SEPARATED
1987 1950 T89F TOSS THOT ZONT—TO0S TOON SOTTO F020
Time (year)
decreases and then it becomes constant.
MOVING PEOPLE
10,000 |
7,500
5,000
2,500
; EPS rt
—
1987 Tass T998 2008 TOIT 2017
Time (year)
experienced people willing to separeto:chg in price people
inexperienced people not willing to separate: cha in price people
inexperienced people willing to separate: chg in pice people
fexperienced people not willing to separate: chg in rice s+» people
Chart 13 A: Fraction bumed and separated (Giving price
incentives)
WASTE SEPARATED
800,000 kg/year |*—F~
© kgiyear
1987 1393 Ta9g 2008 2017 OI
Time (year)
2M kgiyear
60,000 kglyear
14M kgfyear
30,000 kglyear
total amount appropriately separated : chg in price. ys kg/year
total amount recyclable material : chg in price gee = = = kglyear
total amount inappropriately separated : chg in price pq pee gly car
Charts 13 B-C: Impact and outcome of giving price incentives
“Growing number of recycling streams”
A growing number of recycling streams both creates more recycling opportunities and also increases
the amount of recyclable material.
Charts 14 A-C illustrate a slight increase in the
fraction recycled between 1987 and 1993. And
after 1993 the fraction starts to decrease due to a
sharp increase in people becoming unwilling to
separate. The amount appropriately recycled falls
below its initial value and the amount
inappropriately recycled will increase as the
number of people unwilling to separate increases.
The compliance gap increases. Here again, the
model parameters operate near the tipping point.
A higher maximal acceptable time for recycling,
could lead to an opposite policy outcome.
MOVING PEOPLE
1987 135s 199m 2008 TOIT 2017
Time (year)
‘experienced people willing to separate: chg in streams —p— pee prope
inexperienced people not willing to separata = chg in steams ——=—~—e Poople
inexperienced people willing to separate chginstrcams =——5——s———5-—— people
experienced people not willing to separate chq in streams +++ people
REFERENCE FRACTION SEPARATED
1987 1990 1993 1996 1999 2002 2008 2008 2011 2014 2017 2020
Time (year)
three median smooth fraction burned :chg instreams, gg Dal
three median smooth fraction separated: chginstreams. ———> > > > mal
fraction for burning : chg in steams Daal
fraction separated» chg instream Daal
Charts 14 A: Fraction bumed and separated (Increasing
number of recycling streams)
WASTE SEPARATED
AM kgiyear
0.2 kglyear
2M. kgiyear
0.1 kglyear 7
se a es emcee
kglyear
kgiyear |
1987 Tass 7399 2005, 201 2017
Time (year)
total amount appropriately separated : chg in streams —s 12 kgiyear
total amount recyclable material: chg in streams 222 aa kgiyear
total amount inappropriately separated : chg in streams ss kgiyear
Charts 14 B-C Impact and outcome of growing number of recycling streams
Policy lessons learnt
The overview of the policy- experiments (see Table 4) shows that the combinations of price policy
and offering more recycling streams gives the best outcome conceming the fraction separated.
However, this policy results also in two unintended consequences. On the one hand the price
structure (garbage bag charge) leads to a deficit in the solid waste management budget. This deficit
is a result of an intemal feedback structure that is explained in the dynamic hypothesis (Figure 5),
but not yet captured in this model version. Furthermore, once a price incentive is shown to create a
clear gain for citizens to separate, a further increase in the price does not show any remarkable effect
on the fraction separated. Moreover, it even worsen the quality of the separated material. Citizens
not willing to separate might try to avoid the disposal cost by putting un-recyclable material in the
recycling streams. However, this effect will be attenuated since more people will become willing to
separate.
A further insight from simulating altemative policy options is related to the question: Which cues do
we use to observe the policy performance? A glimpse on the fraction separated of the three
altemative policies could tell us that there is only a small difference (the fraction separated stays
relatively low in all three altemative policy- experiments , between 23-38%). But the simulation runs
of the models highlight that there are important differences in the impact and outcome of each
policy. Only the experiment “buming gets expensive” will show a robust policy impact getting
people motivated to participate in recycling programs and improving the outcome. Conversely, due
to the tipping point the two experiments “do nothing”, and “more tasks” can either motivate /
disappoint or overwhelm the citizens, resulting either in a worse outcome (with less material
appropriately separated) or in a slightly better outcome (see Table 4).
Policies |Combination {Do nothing | Burning get | More tasks
Price, streams | No prices, no expensive More streams
chg streams | Price
(Tipping point!) (Tipping point!)
Results
Reference Frac separated Frac separated | Fracseparated | Frac separated
mode
AA >s a >»
—~J =A
Impact Citizens get Citizens get Citizens get Citizens are
Citizens motivated disappointed motivated overwhelmed
Outcome =| Amountapp. Amount app. Amount app. Amount app.
Amount separated aa separated WA separated ” separated & A
and quality | Quality Quality 4 [Quality \% | Quality wa
{and deficit& (and resistance)
resistance)
Table 4: Overview policy tests
Playing with the model shows us that the outcome of an increase in the number of recycling streams
depends on the number of people willing to separate. If there is already a certain social norm to
separate in a community, the effect of an increase in the number of recycling streams will increase
the amount of separated material. Conversely, in a community with a low social nom to separate, an
increase in the number of recycling streams can overwhelm the people, resulting in even less
appropriately separated material. The effect of an increase in the number of recycling streams
depends not only on the social nom to separate, but also on the overall willingness to invest time in
separation. The upper limit indicates a maximal capacity to separate. This interpretation of the
observed tipping point in the model behavior suggest that in the long run a successful separation-
strategy has to be sensitive to the number of recycling streams that are offered. The important
information is the potential capacity of the citizens to separate but also the potential capacity to
separate in the recycling sector. The latter will depend on the market development and the former on
the social nom to separate and the maximal willingness to invest time in separation activities. These
insights are in line with findings of an entropy theoretical discussion of waste management (UIli-
Beer 2000) but also with some insights from computer based simulations of theories about
environmental behavior (Mosler, Gutscher et al. 1996).
In sum the model gives evidence of the superiority of a mixed strategy, motivating citizens to
participate and offering adequate opportunities. While only trying to motivate citizens, contextual
factures could constrain their intention to separate. Similarly, if the focus is only on improving
contextual factors, personal factors (such as a low willingness to spend time on separating) could
inhibit the success of the policy initiative. However, the side effects of an extrinsic motivation
(giving price incentives that results in higher impurity) can be harmful for the overall recycling
initiative. A high impurity can become a trap for the recycling market. Therefore the challenge for
local authorities would be to find policy strategies that helps to build up an intrinsic motivation to
separate.
A further observation is that in all policies, there remains a gap between the amount of the possible
recyclable material and the amount appropriately separated. The width of the gap can be interpreted
as the compliance to separate. It depends not only on the number of people willing to separate and
on other factors such as leaming processes, changes in habits, and the design of the products but also
on the indolence of people. The simulation nuns illustrate that there will never be a 100% separation:
compliance.
The insights about a maximal separation compliance and a maximal separation capacity gives
evidence that structural elements will constrain the overall possible propensity to separate at the
local level.
Conclusions
This modeling project is strongly guided by a feedback perspective on human behavior and policy.
This perspective influences the model design in two ways.
Firstly, it helps to focus on “hidden” personal factors in a system. The theory emphasizes the
existence of such factors, and helps to reflect on the nature of those concepts. It gives an idea how
they affect the system and helps to design them in the model. Disposal habits of a group of people
can be measured in the amount of appropriately separated material. An overall propensity to
separate is determined by different behavioral habits of groups of people. Observed changes in the
propensity to separate indicate changes in behavioral habits of people, leading to a differently
structured society with new social noms. Furthermore, this line of thinking sharpens the focus on
processes, explaining
¢ how contextual factors and personal factors interact with each other and
* how they influence the decision points and
¢ how and where they affect the state of the system.
Likewise, the System Dynamics modeling approach underscores this thinking discipline by focusing
on the “physics of the systems”. Conversely the decision niles in the model are only based on
available information about the state of the systems, representing the theory that that the rate of
change can only be controlled by perceived cues.
Secondly, the feedback theory about human behavior and public policy also guides the search for
possible intervention points in the system. With the picture in mind that interventions can affect both
personal and contextual factors, different intervention strategies can be designed. Policy
interventions aiming to motivate citizens to participate in separation activities, will be different from
policy interventions that aim to improve separation habits of inexperienced but willing people.
While the theory illustrates different intervention points the System Dynamics model helps to
differentiate those and also gives an understanding about the dynamics and effectiveness of
interventions.
To conclude, complementary insights can be gained in the process of applying the theory in a
System Dynamic model.
However, these are only first conclusions. The model has to be developed further and has to endure
and pass different tests of logical coherence, and structural and behavioral correspondence. To
become useful for the local decision makers, they have to gain confidence in the model; the structure
and the behavior must make sense to them. Furthermore the model must be found useful to address
the problems at hand. In order to address these requirements, further workshops with experts in the
area of solid waste management will be arranged. A first feedback from the gatekeeper and
representatives of the model audience" was very encouraging. It was related to the relevance of the
problem statement and the model assumptions.
References
Bauverwaltung Gemeinde X (1997). Mengenstatistik Kehricht, Bauverwaltung.
Black, J. S., P. C. Stem, et al. (1985). “Personal and contextual influences on household behavior.”
Joumal of applied psychology 70(3-21).
BUWAL (2001). Kunststoffrecycling in der Schweiz. B em, Bundesamt fiir Umwelt, Wald und
Landschaft.
Chung, I. J. (1992). Govemment Regulation of Market Information As a Public Policy Tool: The
Dynamics of Waste Recycling Market Development. Departement of Public
Administration. Albany, Nelson A. Rockefeller College of Public A ffairs and Policy: 273.
4 Foran explanation of different roles in group model building see (Richardson and A ndersen 1995)
Cialdini, R. B., R. R. Reno, et al. (1990). “A focus theory of normative conduct: Recycling the
concept of noms to reduce littering in public places.” Journal of personality and social
psychology 58(6): 1015-1026.
Dahlstrand, U. and A. Biel (1997). “Pro- environmental habits: Propensity levels in behvioral
change.” Joumal of applied social psychology 27: 588-601.
Duggan, E. (2002). Costs, change cloud recycling's future. Times Union. Albany, New Y ork: 1, A9.
Einwohnergemeinde X (1973 - 2000). Abfallreglemente, Einwohnergemeinde X .
Einwohnergemeinde X (1973 - 2000). Gebithrentarife zum A bfallreglemente, Einwohnergemeinde
Xx.
Forrester, J. W. (1961). Industrial Dynamics. Cambridge, MIT Press.
Fonrester, J. W. (1994). Policies, decisions, and information sources for modeling. Modeling for
leaming organizations. J. D. W. Morecroft and J. D. Sterman. Portland, Productivity Press.
Gessner, W. (1996). Der lange Arm des Fortschrittes. Umweltproblem Mensch.
Humanwissenschaftliche Zugange zu umweltverantwortlichem Handeln. R. Kaufmann-
Hayoz and A. Di Giulio. Bem, Haupt: 263-299.
Gessner, W. and S. Bruppacher (1999). Restriktionen individuellen umweltverantwortlichen
Handelns. Umweltgerechtes Handeln. Barieren und Briicken. V. Linneweber and E. Kals.
Basel, Birkhauser: 21-47.
GSA (1987 - 2001). Abfallerhebung, A mt fiir Gewasserschutz und A bfallwirtschaft des K antons
Bem.
Hopper, J. R. and J. M. Cart Niesen (1991). “Recycling as altruistic behavior. Normative and
behavioral strategies to expand participation in a community recycling program.”
Environment and Behavior 23(2): 195-220.
Joos, W., V. Carabias, et al. (2002). Ansatze zur Erhebung, Bewertung und V erbesserung der
Sozialvertraglichkeit in der A bfallwirtschaft. Winterthur, Zircher Hochschule Winterthur.
Kaufmann- Hayoz, R., C. Battig, et al. (2001). A Typology of Tools for Building Sustainable
Strategies. Changing Things - Moving People. R. Kaufmann-Hayoz and H. Gutscher.
Basel, Birkhauser: 33-108.
Kaufmann- Hayoz, R. and H. Gutscher, Eds. (2001). Changing Things - Moving People. Strategies
for Promoting Sustainable Development at the Local Level. Basel, Birkhauser.
Kaufmann- Hayoz, R. and H. Gutscher(2001a). Transformation toward Sustainability: An
Interdisciplinary, A ctor-Oriented Perspective. Changing Things - Moving People. R.
Kaufmann- Hayoz and H. Gutscher. Basel, Birkhauser: 19-26.
Latané, B. (1981). “The psychology of social impact.” American psychologist 86(4): 343-356.
Ludwig, C., S. Hellweg, et al., Eds. (2003). Municipal solid waste management. Strategies and
Technologies for Sustainable Solutions. Berlin, Springer.
Morecroft, J. D. W. (1988). “System dynamics and microwonlds for policymakers.” European
Joumal of Operational Research 35(1988): 301-320.
Mosler, H.-J. (2000). Computersimulation sozialpsychologischer Theorien. Weinheim, Psychologie
Vedlags Union.
Mosler, H.-J., H. Gutscher, et al. (1996). Kollektive V eranderungen zu umweltverantwortlichem
Handeln. Umweltproblem Mensch. R. Kaufamm-Hayoz and A. D. Giulio. Bem, Haupt:
237-260.
Oates, E. W. (1990). Decentralization of the Public Sector. An Overview. Decentralication Local
Govemments, and Markets. Towards a Post-Welfare Agenda. J. R. Bennett. Oxford,
Clarendon Press.
OECD (2000). OECD in Figures; Statistics of the Member Countries. Paris, OECD Publications.
Powers, W. T. (1973). Behavior: The Control of Perception. Hawthome, NY, Aldine de Gruyter.
Randers, J. (1996). Guidelines for Model Conceptualization. Modelling for Management II. G. P.
Richardson, Pegasus: 117 - 139.
Reno, R. R., R. B. Cialdini, et al. (1993). “The transsituational influence of social noms.” Joumal of
personality and social psychology 64(1): 104112.
Richardson, G. P. (1991). Feedback Thought. Philadelphia, University of Pennsylvania Press.
Richardson, G. P., Ed. (1996). Modelling for Management ll: Simulation in Support of Systems
Thinking. Vermont, Dartmouth Publishing Company.
Richardson, G. P. and D. F. Andersen (1995). “Teamwork in group model building.” System
Dynamics Review 11(2): 113-137.
Richardson, G. P. and A. L. Pugh (1981). Introduction to System Dynamics - Modeling with
Dynamo. Cambridge, Productivity Press.
Saeed, K. (1992). “Slicing a complex problem for system dynamics modeling.” System Dynamics
Review 8(3): 251-261.
Schwaninger, M. (1997). “Integrative Systems Methodology: Heuristic for Requisite V ariety.”
Intemational Transaction in Operational Research 4(2): 109-123.
Sterman, J. D. (2000). Business Dynamics. Systems Thinking and Modeling for a Complex World.
Boston, Irwin McGraw-Hill.
Stem, P. C. (1999). “Information, incentives, and proenvironmental consumer behavior.” Joumal of
consumer policy 22: 461-478.
Uli- Beer, S. (2000). Unser A bfall - der entwertete Rohstoff der kommenden Generationen?
Entropietheoretische Betrachtung angewandter dkonomischer Modelle und Instrumente im
Bereich der A bfallwirtschaft (Lizentiatsarbeit). Interfakultare K oordinationsstelle fiir
Allgemeine Okologie. Schriftenreihe Studentische A rbeiten Nr. 15. Bem, Universitat Bem.
http://www.epa.gov./epaoswer/non: hw/muncpl/facts.him (US Environmental Protection A gency).
http://www. IP- Waste.unibe.ch/public/A bschlussband/inhaltsverzeichnis.html.
http://www snf.ch/SPP_Umwelt/overview.html.
http:/;www.umwelt- schweiz.ch/buwal/eng/medien/umweltbericht/druck/index. html (Swiss A gency
for the Environment, Forest and Landscape).
EXTERNAL STRUCTURE:
Natural and cultural factors
and processes
Physical
Socio-economic
| Socio-cuttural | Legal -administrative
Objective options and
constraints
ACON
INTERNAL STRUCTURE:
a Goals
Psycho -physiological factors
and processes:
Intention
——— Perceived reality
PERCEPTION
Figure 2: A simple feedback theory about human behavior and public policy
(Kaufmann -Hayoz etal, 2001:82)
Perception of pollution
The emnironrert
*Pollution) |Natural resources
7
° recycling streams
ze ~— | Fbacklog
Gi vceee Soststrum mugs for different
iaceon 2 Impurity (upOu cies
environment @
3 .
Supply of The household waste
Production/ ree. separation sector
Supply in products *amount of separated and
sector burning material
*purity of
separated waste
*relative price
Amount and
quality of
collected
material
The local swm sector
*price
capacity to collect
separated waste
Indicated number of
Recycled raw material
Number of
recycling streams
and relative prices
The household
decision sector
Wumber people willing /
not willing to separate
Number of
recycling streams
Prices and public
services
*N umber people
willing/ not willing
to separate
*willingness to
spend time
*willingness
to spend money
*learning
social norm to
separate
*perception delays
knowledge
Local separation sector
Recycling sector
Feapacity
Amount and
L
backlog ; ; capacit
*price Price for recycling Price for burning backlog _]
Amount
msw for burning
Incineration sector
*number of recycling streams
Amount of separated or
—=—<
recycled raw material for burning
Impact on environment
*price
—
AOC O ROO eRe EAE HERO EOE HEH OREO OSE OEE SEES EEOEOEEOEOSEES ESE OE HEE EEH ES
Figure 9: Overall model structure
fractional rate
becoming unwilling
\.
people willing to separate
disappointed ep
people not willing to separalke
time to learn disappointed
experienced inexperienced iep inexperienced
people not
people willing to people willing fat mater
separate wiep getting | to separate iep getting separate
7 7 experienced motivated
il
fe
time to forget,
nwep losing
experience
y
experienced
people not willing
to separate
fraction willing to
epgetting remotivated
al
Separate fractional rate
becoming willing
perceived
social norm ractiona heftect of
separating <effect o
separating
time cost
cost>
separating>
Figure 10: Changes in citizen’s willingness to separate
(ep: experienced people, iep: inexperienced people, wiep: willing inexperienced people, nwep: not willing experienced people)
multiplier for
Four action choices recyclable material
from number of
recycling streams 0
<experienced people
willing to separate>
<waste per capita
per year wep>
waste generated
by wep per year
actual recyclable
material per person 0
actual total amount actual possible
nonrecyclable material <= recyclable amount
from wep per year
from wep per year
Bl B2 A2 Al
nonrecyclable inappropriately recyclable disposed
lately
disposed for separated by we appropr
burning by wep P i RB for burning by wep i separated by wep
per year per year pee year
per year
4 WS Zz.
<experienced people
willing to separate>
Figure 11: Action choices for disposing the waste (wep: willing experienced people)
The behavioral variables (indicated by diamonds) represent disposal habits. They measure the normal amount inappropriately separated (B2-waste) and the normal
amount appropriately separated (A 1-waste). They also determine both counterparts: the amountrecyclable disposed for burning (A 2-waste) and the non-recyclable
disposed for burning (B1-waste).
Go Back
Recycling sector
~y\ Production / Supply Sector
—_~
number Tecyeling
reycling capacity of
streams recycling
industry
+
? turfs
+
Tecycled raw
material waiting
to be turned into
goods
+ +
Capacity tor
recycled raw
material in
production
industrie
<——$>>,
+
taal j
separated material
burning
Separated
A recyclable
i" +
Tecyclable
material in
goods waiting to
be separated frotia waste for
burning
+=
+
Consumption,
separating and
collecting capacity
did
aE
in localities
Local separation sector
Figure 12: Conceptual overview: Effects of demand and backlogs on capacity developments
Back to th
CC BY-NC-SA 4.0